Theo Honohan on Sat, 14 Mar 2009 19:12:45 -0400 (EDT)

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: <nettime> Cybernetics and the Internet

2009/3/8 Kevin Hamilton <>:
>Brian Holmes wrote:
>> But I would still make a "first/second-order distinction. A
>> Wiener- type machine is always aimed at "changing some singular
>> factor in its environment: that is its purpose, the "end-goal, the
>> "teleology." The second-order systems approach tries to change the
>> parameters of the entire envoronment and not just one factor within
>> it. At the extreme limit, that means changing the parameters of the
>> world economy, or even of world society.
> Yes, I want to keep this question of teleology in place - it's how we
> keep from mistaking a method for a moral, since the same method can
> be placed in the service of very different morals. I do see a clear
> enough distinction between first and second-order cybernetics, but
> where it gets fuzzy for me is when "changing the parameters of the
> entire environment" is for some a means of achieving a clear static
> end, a singular factor such as world domination. Then something that
> started out very second-order looks very first-order.

If a cybernetic view is supposed to be more realistic than the alleged
"god delusions" of self-organization, this idea of "changing the
parameters of the entire environment" is a steep requirement for it to
satisfy without itself becoming a god delusion.

The distinction between first and second order cybernetics is clear,
but to valorize it is to accept the narcissism of subjectivity.  The
image currently displayed on the Wikipedia page for  second order
which derives from
(due to Bateson and Mead),
expresses this perfectly.

In the system the diagram portrays, second order cybernetics replaces
the (supposedly reductionist) engineer of first order cybernetics with
the term "feedback", while Wiener, Bateson and Mead appear as fully
conscious observers at a higher level of abstraction. The elitism and
narcissism implicit in this depiction is more than just amusing, it's
intrinsic to the fatal direction of the cybernetic project.

Any mathematician or computer scientist would point out that Wiener,
Bateson and Mead, for all their enlightened holism, could equally
be replaced with the term "feedback" in a system observed by
others--Holmes, Stalder and Hamilton, say. This is not a discipline
with two distinct modalities, first order and second order, it's a
unitary activity which has a self-referential structure.

IMO the decision to talk about "second order cybernetics", rather
than just publish better papers about the generality of cybernetics,
was misconceived and inelegant. Though I'm sure I have read much less
about this than Brian or many of the rest of you, there is an easily
recognizable posturing and "hand-waving" quality to so much of the
cybernetics work that this choice is commensurate with the general
tone of things.

Computer science approached analogous issues of generality when
the question of considering functions as "first class citizens"
(Christopher Strachey) arose in the 1960s. The resulting terminology
and equipment now in use deals in terms of "first class objects" and
"higher order functions" rather than fixing the degree of reference
at any particular number of levels of remove. (The inelegance of
using numbers other than "one" and "many" should be familiar to
philosophers; it's a red flag. Number theory has many complexities and
depths but they are rarely explicable in any satisfactory way, being
more chaotic natural artifacts than objects of logic)

Perhaps that's a bit of a rant but my point is that although the
move to second order was an acknowledgement that the engineer could
be considered as an entity in the scope of the model, it necessarily
suppressed the fact that Wiener, Bateson and Mead were, and always had
been, entities in the scope of the model. Perhaps not accounted for by
the model, but nevertheless passively present. Capable, for example,
of sabotage by throwing their clogs into the modelled mechanism. The
kind of schizoid distancing of the modeller from the action obscures
the fact that the system (reality) has always included them, has
always been universal and closed.

So the possibility of "changing the parameters of the entire
environment" sounds intriguing, but the idea of achieving it
cybernetically requires (an almost) godlike power, by definition.
On the other hand, material evidence implies that reality involves
a ineluctable process of self-organization, to the extent that free
will is excluded. Of course, metaphysically, both views could be
simultaneously valid, and at the same time neither tells us how to

The hubris, the grandiose scope, the collective psychopathology(!),
particularly the narcissism, of the cybernetic movement, combined with
a failure to find practically applicable techniques, is surely what
killed it, rather than narrow-mindedness in the world of scientific

The idea that some system, some individual, is in control is
fundamentally narcissistic, and thus corresponds closely with the
imaginary of hierarchical political culture, of course. But of course
it's not the only reality.

Bifo Berardi suggested at his talk in London a week or two ago that
the primary political activity now should be therapy. Perhaps dealing
with the equally attractive, opposed drives towards cybernetic mastery
and self-organizing lethargy would be a good place to start.


#  distributed via <nettime>: no commercial use without permission
#  <nettime>  is a moderated mailing list for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info:
#  archive: contact: